Local control of thoracic cancer by fractionated radiation therapy, with or without chemotherapy, is quite poor. There are emerging data for non-small cell lung cancer (NSCLC) that show higher doses (>70 Gy) delivered with conformal radiation therapy (CRT) may provide better local control than lower doses (<70 Gy). The radiation beam size is routinely increased beyond the projected tumor size to account for breathing motion and patient set-up uncertainty. A consequence of this is an increased normal lung irradiated volume, which is the main limiting factor to dose escalation. Reducing the irradiated volume will require reductions in the planning and delivery geometric uncertainties. While there have been many individual solutions for simulation techniques and delivery technology, there is not yet a solution that integrates high temporal measurements of the tumor position with low temporal, but high spatial resolution measurements of tumors and normal structures that can be applicable to the majority of locally advanced lung cancer patients. We propose the integration of real-time tumor tracking with cross-sectional imaging into a radiation therapy treatment planning system that can measure the effects of geometric uncertainty with high precision. PUBLIC HEALTH RELEVANCE: Lung cancer is the most common cause of cancer death in the United States. This study proposes a method of targeting lung cancer with radiation more precisely. If this can be done, one can increase radiation dose to the tumor, resulting in improved control of the tumor.